MLP Soup: AMTP May Be Better Mousetrap

If I had to pick a winner for the “Most Annoying FAQ” to our analytics hotline, it would be the question “What’s the best MLP ETF.”

As Paul Baiocchi and I wrote in our whitepaper on MLPs , for most investors the answer is going to be “None of the Above,” and for one very good reason – taxes.

The whole point of investing in master limited partnerships has been, for decades, tax deferral. While there are plenty of reasons one might want to be in the oil and gas transport business—the business of the vast majority of MLPs—the really interesting thing about MLPs is the “P” part, namely: Partnership.

Because these big pipeline MLPs have enormous investments in infrastructure to write down, but consistent income, they have been able to pass back out to their partnership stakeholders both a big tax write-offs and a big chunk of cash on regular basis. That means regular income that gets treated as a “return of capital” by the Internal Revenue Service.

That’s tax-free money that reduces your basis in the MLP itself. Once the basis runs out—generally after many, many years—you’re of course potentially left with a zero-basis investment and all future income would be taxed just as ordinary income. But at that point, the logic often goes, you can gift the original MLP shares off to your favorite charity (or grandson) and call it a day.

All of those tax shenanigans disappear in the exchange-traded product versions, where you’re stuck with a corporate shell owning the MLPs for you (and paying taxes before they pay you) or, in an ETN, tracking an index (in which case you get ordinary income coupon payments and long-term capital gains, just like a plain old stock investment.)

Still, despite this loss of favored tax treatment, there seems to be tremendous investor interest.

The big daddy in the space, The Alerian MLP ETF (AMLP) , now has nearly $7 billion under management, despite the super-unfavorable and complex tax-treatment of its corporate structure, and the JP Morgan Alerian MLP ETN (AMJ) , has $5.5 billion despite having been closed for creations for more than a year.

Our answer on “what to pick” has generally leaned towards the admittedly broken AMJ, because at least it avoided the corporate tax treatment issues, and being an ETN, promised good tracking of the underlying index.

In March, however, a new kid came to town—the Barclays ETN+ Select MLP ETN (ATMP) .

To put it in context, here’s how it’s performed on a total return basis since it launched back in March, vs. it’s main competitors.

View gallery


I have a soft spot for the guys putting out the ETN+ products. They came out with a clever and elegant solution to the daily-rebalancing leveraged and inverse problem, and in some ways, they’ve done a similar thing here. The actual index is clever in two ways.

First, the index caps its holdings in individual MLPs to 8 percent. That keeps Kinder Morgan from taking over the index. Further, the index includes General Partnerships—folks that actually manage pipelines, not just own them.

GPs generally don’t have the same crazy tax-dodge quality as the LPs, but since all those issues are irrelevant in an ETN, it’s actually smart diversification, and makes the index a better sector play on the business, and less a tax strategy.

GPs included in the index are meanwhile capped at 4 percent each. Of course, some of the GPs they own are essentially the same company as the LPs, but in general, I think it makes sense.

The whole portfolio is rebalanced regularly, which is also prudent indexing best-practice, in my opinion. It takes money “off the table” and reallocates it to underperforming securities, and long term, I think that’s a smart move inside sectors.

Another smarty-pants idea: The version of the index the ETN actually tracks is based on the volume-weighted average prices of each of these firms. That’s smart, because it ensures that the prices for the index are based on actual trading.




MLPs often trade at much lower volumes than their market cap might suggest, and if Barclays has to be in the market buying MLPs to offset the ETNs obligations, they don’t want to be in the position of buying at one price while at the same time promising a different price in the index. Basing their promise to investors on volume-weighted average price (VWAP) matches their liability to their assets. Again, it’s a sensible move.

So, it’s a slightly better mousetrap on the index side. What about the actual structure?

As ETNs go, the Barclays ETN+ documents are relatively clean and straightforward. While the math is a bit cumbersome, the ETN is promising to pay out a coupon payment in proportion to the implied dividends from the index—to the extent they exceed fees.

Like many ETNs, the way that fee—a fairly steep 95 basis points—is assessed isn’t all that investor friendly. It’s done off the price level of the index, no matter what, not off the notional value of the ETN.

In other words, if the fair value of the ETN was $25 for the rest of history, and no dividends were ever paid, that 95 basis points would be charged on the $25 level, year after year, never taking into account that your actual investment had already been degraded by fees already paid.

That sucks, but it’s not all that unusual in an ETN.

As an issuer, Barclays is about as safe as an ETN counterparty can be. (It gets a “low” in our counterparty risk scoring). There’s a 12.5 basis point redemption fee, which again, isn’t unusual, and simply means you’d expect it to trade at a slight discount should there be a rush to the exists by investors for some reason.

So all in all, Barclays’ ATMP is an entirely ownable exposure to an interesting take on MLPs. The problem is tracking your investment.

The data on ATMP is, frankly, terrible. Here’s how the ETN looks when you track it against the two versions of the index Barclays publishes on Bloomberg.

View gallery


If you think that’s ATMP crushing it’s index, your wrong.

Neither version—the volume weighted or the regular index—is published as a total return series, so the roughly 4 percent annual dividends the underlying MLPs are paying out is simply ignored.

Normally I’d call them out for sandbagging – deliberately picking a non-total-return index to make themselves look better, but even their own website publishes the wrong data. Here’s the chart from the Barclays ETN+ website:

View gallery


And that’s the next problem. Good luck even finding information on this product (it’s at, but you don’t get that on the first page of a Google search). Once you do find the website, there’s no performance table, no list of current holdings in the index. The documentation is there, but it’s all still from the fund’s launch in March.

If you want to know what the current index weights are, you might expect you could go to – after all, Atlantic Trust supposedly came up with the index. But have fun even finding acknowledgement of their involvement with the product.

As for index updates – Barclays has put out a press release or two noting that the quarterly rebalances occurred, but I wasn’t able to find them anywhere but the PR wires – not on the fund website, not on the index website.

Even our own analytics product struggles with this kind of data issue. In general, we’ll make the call to simply not include data if we think it’s not good, and so you’ll see a lot missing from AMTP’s pages here.

It’s a shame, honestly.

The next stage of growth in ETFs is going to come from retail investors or financial advisors who, by definition, have been skeptical of ETFs to date, but are coming to realize their incredible benefits.

But if they stumble across products like AMTP – better mousetraps that live in informational black holes – they may never make the jump.


Permalink | © Copyright 2013 IndexUniverse LLC. All rights reserved
View Comments